Back to Talks

Open Geospatial Machine Learning

Kevin Stofan DataRobot

Audience level: Intermediate
Topic area: Modeling

Description

We will guide attendees through the entire geospatial machine learning workflow. Attendees will be be exposed to a variety of open source tools used to process, model, and visualize geospatial data. This workshop will focus on concepts unique to handling geospatial data such as spatial autocorrelation, lagged spatial features, and spatial partitioning.

Abstract:

Intro and Prelims

  • Installation

  • Agenda

  • Background

Geospatial Data Formats and I/O

  • Geospatial Data Models

  • GDAL and OGR

  • File-based Formats

Exploratory Spatial Data Analysis (ESDA)

  • Python Spatial Abstraction Library (PySAL)

  • Spatial Weights Matrix

  • Spatial Autocorrelation

Smooth, Regionalization, and Neighborhood Analysis

  • Spatial Smoothing

  • Regionalization

  • Neighborhood Analysis

Geospatial Feature Engineering

  • Geometry-based Features

  • Topologically-based Features

  • Set theoretic Features

Geospatial Feature Enrichment

  • Joins

  • Areal Weighting/Dasymetric Mapping

  • Zonal Statistics

Spatial Econometrics

  • Spatial Regression

  • Spatial Lag

  • Spatial Error

Machine Learning with Geospatial (GBM Example)

  • Spatial Partitioning

  • Spatial Validation/CV

  • Spatial Partial Dependence